I don't think this is a serious test. It's just an art piece to contrast different LLMs taking on the same task, and against themselves since it updates every minute. One minute one of the results was really good for me and the next minute it was very, very bad.